CN107045367A - A kind of greenhouse multiple-factor coordinates energy-conserving and optimizing control method - Google Patents

Abstract

The invention relates to a multi-factor coordinated energy-saving optimization control method for a greenhouse environment, comprising the following steps: 1) setting the expected daily average temperature of crops in each growth period, and obtaining weather forecast data for the next seven days; 2) estimating the ventilation system of the greenhouse State; 3) according to step 1) and step 2) use multi-factor coordination algorithm to set the set value of each environmental factor in the greenhouse, said environmental factor includes temperature, humidity, light radiation intensity and carbon dioxide concentration; 4) obtain environmental factor real-time value, and adjust the corresponding actuators in the greenhouse according to the set values of the environmental factors. Compared with the existing technology, the present invention uses the physiological characteristics of crops such as accumulated temperature to effectively reduce the energy consumption of the greenhouse. At the same time, the corresponding strategy is adopted to coordinate the secondary factors of the greenhouse environment to the main factor of temperature, which not only ensures the precise regulation of the greenhouse and efficiency, avoid unnecessary waste of energy consumption, save the cost of regulation, and ensure accurate and effective control.

Description

A multi-factor coordinated energy-saving optimization control method for greenhouse environment

technical field

The invention relates to the technical field of agricultural environment control, in particular to a multi-factor coordinated energy-saving optimization control method for a greenhouse environment.

Background technique

Greenhouses are the infrastructure for facility agriculture and factory agriculture. Greenhouse environment control is based on making full use of natural resources, by changing greenhouse environmental factors such as temperature, humidity, light, carbon dioxide, etc. to meet the quantitative requirements of the greenhouse climate in each growth period of crop growth, and realizing the control of greenhouse climate data through the computer control system. The acquisition and processing of the greenhouse, and the corresponding automatic control algorithm, realize the regulation of the greenhouse ventilation system (skylight and side window), sunshade system, heat preservation system, heating system, cooling system, carbon dioxide release system and other actuators, so as to realize the control of the greenhouse The automatic control of climate creates indoor climate conditions (also known as microclimate) suitable for crop growth. This technology is an important means to improve the yield and quality of greenhouse crops for large-scale factory production, and it is also an important way to regulate the time of crops to market.

After more than 30 years of development, the greenhouse environment control has experienced a development process from only adopting simple winter heat preservation measures to controlling multiple conditions required for plant growth. However, compared with developed countries, the overall efficiency of my country's greenhouse production is still low. The reason is that there is a lack of effective control of greenhouse energy consumption in China. Although the indoor environment can be precisely controlled to increase production, the cost of energy consumption is too high and the economic benefits are low, failing to really increase farmers' income. The high energy consumption of greenhouse production is mainly due to the following reasons:

1. The setting value of the greenhouse environmental factors is unreasonable. The unreasonable temperature set point stems from the unreasonable treatment of the time scale of greenhouse management and process control. Taking cherry tomato as an example, the entire growth cycle of the crop can be as long as 300 days, and for indoor environmental process control, the time unit is usually "minutes" and "hours". Some of the current greenhouse environment control algorithms one-sidedly emphasize the energy consumption management of the entire production cycle and ignore the process control, and cannot achieve precise regulation of the environment in a short time scale, resulting in low yields and affecting farmers’ income; some one-sidedly emphasize short-term Precise control of the time-scale microclimate ignores the management of energy consumption throughout the production cycle. Excessive energy consumption leads to high costs and affects farmers' income. The temperature setpoints in these algorithms are necessarily also unreasonable. Many empirical-based greenhouse environmental control algorithms use static operating point control algorithms, using simple temperature as a single variable, to control the temperature in the greenhouse. This method has the advantages of being simple and easy to operate, and makes full use of the advantage of crops’ adaptability to the environment, but the disadvantages are also obvious. First, when encountering extreme weather such as high temperature or cold wave, the cooling or heating system load will increase. It will greatly increase the energy consumption for greenhouse regulation; the second is that it cannot meet the needs of crops for environmental factors (such as humidity and carbon dioxide concentration) other than temperature.

2. Disharmony between regulation means and environmental factors. There are many environmental factors related to crop growth in the greenhouse, as well as various executive agency systems and various control methods. The regulation effect, control intensity and required energy consumption of various control methods are also different, and sometimes the adjustment results of different methods even conflict with each other. Unreasonable control methods will also increase energy consumption. For example, in winter, the outdoor temperature is low and the humidity in the greenhouse is high. Too low temperature is not conducive to crop growth, and too high humidity may cause crop root rot and other hazards. Considering the temperature, it is necessary to heat the greenhouse, and considering the ventilation, it is necessary to ventilate the greenhouse for dehumidification, and the ventilation will cause a large amount of heat loss and increase the cost of greenhouse production.

Contents of the invention

The purpose of the present invention is to provide a greenhouse environment multi-factor coordinated energy-saving optimization control method in order to overcome the above-mentioned defects in the prior art, which coordinates the secondary factors of the greenhouse environment to the main temperature factor in equivalent coordination, and at the same time ensures that the corresponding Coordination not only ensures the precision and efficiency of greenhouse regulation, but also avoids unnecessary waste of energy consumption and saves the cost of regulation.

The purpose of the present invention can be achieved through the following technical solutions:

A greenhouse environment multi-factor coordinated energy-saving optimization control method, comprising the following steps:

1) Set the expected daily average temperature of crops in each growth period, and obtain weather forecast data for the next seven days;

2) Estimate the status of the greenhouse ventilation system;

3) According to step 1) and step 2), adopt multi-factor coordination algorithm to set the setting values of each environmental factor in the greenhouse, and the environmental factor includes temperature, humidity, light radiation intensity and carbon dioxide concentration;

4) Obtain the real-time value of the environmental factor, and adjust the corresponding actuator in the greenhouse according to the set value of the environmental factor.

In described step 3), the concrete process that temperature is set is:

A1) Determine the weekly average temperature according to the growth period of the weekly crops;

A2) According to the expected daily average temperature and weather forecast data, the rolling optimization method is used to calculate the optimal daily average temperature for the next seven days. The frequency of rolling optimization is once _a day. The performance function J1 used in rolling optimization is:

In the formula, q _tom η _DMFM DM _Har (T _Di ) represents the economic income produced by the crop when the daily average temperature of the i day is T _Di , q _tom represents the unit price of the crop, and η _DMFM represents the conversion factor from fruit dry weight to fruit fresh weight , DM _Har represents the harvested fruit dry matter yield, q _heat Q _heat (T _Di ) represents the cost of heating energy consumption when the average temperature of day i is T _Di , q _heat represents the unit price of heating energy, Q _heat represents the heating energy consumption,

The constraints used in the rolling optimization include seven-day cumulative temperature conditions and indoor temperature upper and lower limit conditions;

A3) The rolling optimization method is used to calculate the hourly average temperature of the day under the optimal daily average temperature constraint, the frequency of rolling optimization is once per hour, and the performance function J2 adopted during rolling optimization is _:

In the formula, T _Hj represents the hourly average temperature of the jth hour;

Constraints used in rolling optimization include daily cumulative temperature conditions, indoor temperature upper and lower limit conditions, daytime average temperature conditions and adjacent hour temperature difference upper limit conditions.

The growth period of the crop includes seedling stage, growth stage and fruit stage.

When obtaining the economic income generated by the crops, the crop yield or fruit dry matter is equivalently distributed to each growth and development stage of the crop growth.

In the step 3), the specific process for setting the carbon dioxide concentration is:

In each hour, based on the current temperature setting value and the light data in the weather forecast data as conditions, the carbon dioxide concentration setting value as the optimization variable, and the goal of maximizing the sum of economic benefits of the control step, optimize and calculate To obtain the carbon dioxide concentration setting value of each control step, the sum of the economic benefits adopted in the optimization process, that is, the performance function J ₃ is:

In the formula, CO _2,k represents the set value of carbon dioxide concentration at the kth control step, q _tom represents the unit price of agricultural products, η _DMFM represents the conversion factor from fruit dry weight to fruit fresh weight, and η _PDM represents the conversion factor of photosynthetic products into dry matter Conversion factor, P represents the total photosynthetic output in the control period, Indicates the carbon dioxide unit, Indicates the amount of carbon dioxide released, and s indicates the total number of control steps in one hour.

The control step is 15 minutes.

In said step 2), the estimated state of the greenhouse ventilation system is specifically:

The temperature value in the weather forecast data is used as the outdoor temperature, the outdoor temperature is compared with the frosting temperature and the ventilation temperature, and the opening degree of the ventilation system is obtained according to the comparison result.

In the step 4), when adjusting and controlling the corresponding actuators in the greenhouse, the principle of adjusting and controlling the mutual coordination between the environmental factors of the greenhouse and the mutual coordination of the control means is taken.

In the step 4), when regulating the greenhouse, it specifically includes temperature control, humidity control, light control, carbon dioxide control and ventilation control.

In the step 4), when the greenhouse is regulated, the weighted linear function T is used to determine the actions of each control means, and the expression of the weighted linear function T is:

T(m _co2 ,m _T ,m _R ,m _H )＝α×F(m _co2set ,m _Tset ,m _Rset ,m _Hset )+β·G(m _co2in ,m _Tin ,m _Rin ,m _Hin )+λ ·H(m _Tout ,m _Rout ,m _Hout ,Fv,P _rain )

In the formula, T represents the specific control means, F represents the parameter value function manually set, G represents the indoor environment parameters, H represents the outdoor environment parameters, α, β, λ represent the corresponding weights respectively; m _co2 represents the amount of carbon dioxide released, m _T represents the target temperature, m _R represents the target radiation amount of light, m _H represents the target humidity, m _co2set represents the set value of indoor carbon dioxide concentration, m _Tset represents the set value of indoor temperature, m _Rset represents the set value of indoor light radiation, m _Hset indicates indoor humidity setting value, m _co2in indicates indoor carbon dioxide concentration, m _Tin indicates indoor temperature, m _Rin indicates indoor light radiation, m _Hin indicates indoor humidity, m _co2out indicates outdoor carbon dioxide concentration, m _Tout indicates outdoor temperature, m _Rout Indicates the amount of outdoor light radiation, m _Hout indicates the outdoor humidity, Fv indicates the outdoor wind speed, and P _rain indicates the outdoor rainfall.

Compared with the prior art, the present invention has the following beneficial effects:

1. The present invention coordinates and designs the methods for obtaining the set values of various greenhouse environmental factors, effectively reducing the energy consumption of the greenhouse and increasing income. On the basis of agricultural experience, the setting values of environmental factors that have a greater impact on plant production are optimized to achieve the purpose of energy saving and income increase.

2. The present invention uses the physiological characteristics of crops such as accumulated temperature, and combines outdoor weather forecasting to predict the outdoor weather. On the premise of ensuring the output of the greenhouse, the starting point is to reduce the energy consumption of the greenhouse, and the method of numerical solution is used to ensure the demand for the accumulated temperature of the crops. , the environmental setting value changes with the outdoor weather, and the resulting dynamic greenhouse environment setting value has a more significant energy-saving effect than the traditional static working point.

3. In terms of control means, the present invention adopts corresponding strategies to equivalently coordinate the secondary factors of the greenhouse environment to the main temperature factors, and at the same time ensure the coordination between the actuators, which not only ensures the accuracy and efficiency of greenhouse control, but also avoids Unnecessary waste of energy consumption saves the cost of regulation and ensures accurate and effective control.

4. The present invention adopts multi-time scale layered hierarchical rolling optimization to optimize the temperature setting value. Among them, rolling optimization is used in the calculation of the seven-day daily average temperature, which eliminates the week and week in the layered hierarchical process. The boundary between them solves the problem of crossing physiological stages in a week; rolling optimization is used to calculate the hourly average temperature, and the deviation is corrected with the actual temperature that has passed the time of the day, which can ensure the optimal daily average temperature.

5. Combined with a week's outdoor weather forecast, through the optimization of the indoor daily average temperature, according to the difference in the outdoor daily average temperature, the accumulated temperature within a week is allocated to each day, thereby saving energy consumption.

6. This application optimizes the concentration of carbon dioxide with a faster rate of change in a control step every hour, and the calculation result is more accurate. According to the outdoor light intensity, the set value of carbon dioxide is determined, and the carbon dioxide is released when the light is strong, so as to promote photosynthesis and avoid waste.

7. The present invention is based on the principle of regulating and controlling the mutual coordination between environmental factors and the mutual coordination of regulating means, and regulates and controls the corresponding executive agencies in the greenhouse. The secondary factors (such as humidity, light, etc.) are all managed to coordinate with the main factors (such as temperature) to establish a coordinated Relational function, so as to change the complex multi-factor control into multi-factor coordinated control with temperature single factor as the main factor, supplemented by feed-forward and feedback control to eliminate some uncertainties brought about by "coordination", and solve the problem of multi-factor greenhouse environment Severely coupled problems, to achieve the purpose of multi-factor control.

Description of drawings

Figure 1 is a structural diagram of the greenhouse environment control system;

Fig. 2 is the greenhouse control flowchart of energy saving;

Fig. 3 is a principle diagram of energy-saving optimization of greenhouse environment setting value based on accumulated temperature;

Fig. 4 is the energy-saving optimization flow chart of the greenhouse environment setting value based on accumulated temperature;

Fig. 5 is the influence diagram of indoor humidity to temperature;

Fig. 6 is the relationship diagram of the influence of outdoor light on temperature;

Fig. 7 is a flow chart of indoor humidity correction ventilation temperature;

Fig. 8 is a flow chart of outdoor light radiation correction for ventilation temperature;

Fig. 9 is a flow chart of the control of the actuator;

Figure 10 is a flow chart of sunroof control.

detailed description

The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is carried out on the premise of the technical solution of the present invention, and detailed implementation and specific operation process are given, but the protection scope of the present invention is not limited to the following embodiments.

This embodiment provides a multi-factor coordinated energy-saving optimization control method for the greenhouse environment. It uses the physiological characteristics of crops such as accumulated temperature, and combines the outdoor weather forecast to predict the outdoor weather. In the solution method, under the premise of ensuring the accumulated temperature requirements of crops, the environmental setting value changes with the outdoor weather. The dynamic greenhouse environment setting value thus obtained has a more significant energy-saving effect than the traditional static working point, and at the same time The corresponding strategy is used to coordinate the secondary factors of the greenhouse environment equivalently to the main temperature factors, and at the same time ensure the coordination between the actuators, which not only ensures the accuracy and efficiency of greenhouse regulation, but also avoids unnecessary waste of energy consumption and saves Regulatory costs.

The greenhouse environment is a multi-factor control system, and there are many control means (executive agencies) to achieve the multi-factor control goal of the greenhouse, as shown in Figure 1. The control method of the present invention simplifies the control of the greenhouse environment on the basis of fully understanding the greenhouse control technology:

a) Arrange circulation fans on the top of the greenhouse to speed up the air circulation and try to make the distribution of the internal environmental parameters of the greenhouse roughly the same;

b) According to the different light radiation intensity, the outdoor weather is divided into four situations: sunny, cloudy, cloudy, and rainy (corresponding to the critical points of light radiation intensity are mR0, mR1, mR2).

As shown in Figure 2, the greenhouse environment multi-factor coordinated energy-saving optimization control method provided by this embodiment includes the following steps:

(1) Initialization. Set the desired average daily temperature for each growing period of the crop, which is derived from gardening experience, and obtain seven-day weather forecast data. Taking tomato after fixed value as an example, the planting experience is summarized in Table 1.

Table 1 Duration days and expected average temperature of each tomato growth period

Seven-day weather forecast data includes daily in a week: outdoor temperature, humidity, light intensity, rainfall, wind speed, wind direction, etc.

The basis for choosing seven days here is: on the one hand, crop growth is directly affected by accumulated temperature, and the calculation time of accumulated temperature should not be too short; on the other hand, if the weather forecast takes too long, its accuracy cannot be guaranteed. The duration of the weather forecast is one week.

(2) Using the multi-factor coordination algorithm to estimate the state of the ventilation system of the greenhouse. The estimation method is to compare the outdoor temperature with the ventilation temperature, see step (407) and FIG. 10 for details.

(3) Set the set values of environmental factors such as indoor temperature, humidity, light radiation intensity, and carbon dioxide concentration. Wherein the temperature and the carbon dioxide concentration are obtained through optimization, which is an innovative point of the present invention. The setting values of humidity and light radiation intensity are set according to gardening experience. Taking tomato as an example, the humidity is 55% at the seedling stage and 70% at the growth stage , 80% in the fruiting period, the light compensation point is 3,000 lux, the saturation point is 70,000 lux, and the suitable range is 40,000-50,000 lux.

The optimization process of temperature and carbon dioxide concentration adopts the multi-time scale layered hierarchical rolling optimization method, and the optimization performance function is the economic benefit function E, which is different in each time scale. The optimization method can use genetic algorithm, particle swarm algorithm and other optimization algorithms.

The economic benefit function E[yuan/m ² ] of greenhouse environment control is:

E＝Q _cropyield- Q _var

In the formula, Q _cropyield [yuan/m ² ] is the economic income generated by greenhouse crops, and Q _var [yuan/m ² ] is the variable expenditure of greenhouse operation.

The economic income Q _cropyield [yuan/m ² ] generated by greenhouse crops can be expressed by the following formula:

Q _cropyield ＝q _tom ×η _DMFM ×DM _Har

In the formula, q _tom [yuan/mg] is the unit price of agricultural products, and η _DMFM is the conversion factor from fruit dry weight (dry matter quality) to fruit fresh weight (fruit yield). According to agricultural experience, its value is generally between 7-20, and DMFM _Har [mg/m ² ] is the harvested fruit dry matter yield.

For fruit and vegetable crops, the fruit begins to appear at a certain stage after the crop enters reproductive growth. In order to ensure that the above-mentioned economic income can be continuously estimated, it is necessary to distribute the crop yield or fruit dry matter equivalently to each growth and development stage of the crop growth. This is also an innovative point of the present invention. For example, referring to the Logistic model of tomato growth and development based on accumulated temperature studied by Qi Weiqiang, the Logistic equivalent distribution Y' of crop fruit dry matter yield DM _Har in the entire production period can be constructed:

In the formula, Y ₀ is the identification parameter related to the total crop yield at the time of harvest, PT is the effective accumulated temperature of the greenhouse, and a and b are the identification coefficients. Refer to the corresponding literature to set the above parameter values Y ₀ = 395.2275, a = 5.5616, b = -0.004023.

Or use the biomass in the Vanthoor model, that is, the carbohydrate C _Buf in the buffer zone or the leaf area index LAI as the consideration index of the equivalent yield.

The variable expenditure Q _var [yuan/m ² ] of greenhouse operation is expressed as:

Q _var = Q _plant + Q _water + Q _CO2 + Q _energy

Among them, Q _plant [yuan/m ² ] represents the cost related to planting, such as the cost of seeds and fertilizers, Q _water [yuan/m ² ] represents the cost of water consumption, and Q _CO2 [yuan/m ² ] represents the increase of carbon dioxide Consumption cost, Q _energy [yuan/m ² ] indicates the cost of heating and cooling energy consumption. The content of the present invention is mainly aimed at the microclimate control of the greenhouse, so irrigation water is not considered, so the planting cost Q _plant and the water cost Q _water can be regarded as constant parameters in the variable cost expenditure, and do not participate in the above optimization process.

The greenhouse energy consumption model is described as follows:

ΔQ＝Q _rad +Q _heat -Q _cond -Q _vent -Q _tran

In the formula, ΔQ is the change of internal energy in the greenhouse, Q _rad [J] is the energy increased by solar radiation, Q _heat [J] is the energy input into the greenhouse by the heating system, Q _cond [J] is the energy lost by the conduction of the greenhouse, Q _vent [J] is the energy lost by greenhouse ventilation, Q _tran [J] is the latent heat consumed by crop transpiration.

According to the knowledge of thermodynamics, the formula for calculating the amount of change in the internal energy of the greenhouse is as follows:

ΔQ＝ρC _p ΔTV _gh

In the formula, ρ is the air density [kg/m ³ ], C _p is the heat capacity of air at constant pressure [J kg ^-1 K ^-1 ], ΔT is the temperature change [K], and V _gh [m ³ ] is the volume of the greenhouse.

Q _rad can be calculated by the following formula:

Q _rad =τ _rad ·I _Glob ·S _gh

In the formula, τ _rad is the solar radiation transmittance, the value is 0.78, I _Glob [W m ^-2 ] is the outdoor light radiation intensity, and S _gh is [m ² ] the area of the greenhouse.

Conduction heat transfer per unit time in the greenhouse and ventilation heat exchange It can be calculated with the following formula ^[30] :

In the formula, U _gh [W m ^-2 K ^-1 ] is the heat loss value, the value of the glass greenhouse is 6.5, A _gh [m ² ] is the surface area of the greenhouse, T _in and T _out [K] are the indoor and outdoor temperature of the greenhouse, respectively. temperature, W is the wind speed factor, its value is affected by the wind speed near the greenhouse, the value is 1-1.075, U _vent is the ventilation control amount, V _gh [m ³ ] is the volume of the greenhouse, E is the air heat transfer coefficient, and the glass greenhouse The value is 1.08.

Since the heat exchange of transpiration is mainly a process between plants and the greenhouse environment, crops and the greenhouse environment are usually considered as a whole in greenhouse regulation, and they are temporarily regarded as fixed parameters.

The carbon dioxide consumption model is described as follows:

The change of carbon dioxide concentration in the greenhouse is the result of the joint action of crop photosynthesis, respiration, ventilation and supplementary carbon dioxide. The equilibrium model is as follows:

In the formula, C _i is the concentration of carbon dioxide in the greenhouse [kg m ^-3 ], h [m] is the height of the greenhouse, C _g is the rate of supplementing carbon dioxide in the greenhouse [kg m ^-2 s ^-1 ], C _i,o is the ventilation The change rate of carbon dioxide caused by [kg m ^-2 s ^-1 ], C _gl represents the rate of carbon dioxide absorbed by photosynthesis in the greenhouse [kg m ^-2 s ^-1 ], C _c,resp and C _s,resp represent crop respiration and Carbon dioxide release rate by soil respiration [kg m ^-2 s ^-1 ].

In most greenhouses, a layer of thermal insulation film will be covered on the ground surface in cold winter, and the respiration of the soil C _s,resp can be ignored; for the needs of thermal insulation and energy saving in winter, the skylight is at a small angle or fully closed most of the time, and the ventilation The carbon dioxide change C _i,o caused can also be ignored; the carbon dioxide change C _gl and C _c,resp caused by photosynthesis and respiration can be calculated with reference to crop physiological models, such as tomato crops can use crop physiological models such as Vanthoor and TOMGRO.

The logical principle and specific steps of step (3) are shown in Figure 3 and Figure 4 respectively, specifically:

(301) Determine the weekly average temperature T _M1 , T _M2 ... T _Mm according to the growing season of the weekly crops. According to long-term horticultural experience, crops can be divided into three growth stages: seedling stage, growth stage, and fruit stage according to different physiological characteristics and different requirements for the environment. Step (1) has set the expected daily average temperature of each growth stage (that is, the optimal daily average temperature) T _N1 , T _N2 , T _N3 , then the weekly average temperature T _M1 , T _M2 ... T _Mm can be obtained from the growth period of the weekly crop. The average temperature of crops at a certain growth stage P Among them, T(t) is the temperature value at any time in the greenhouse during the time period, and P represents the time length of the growth stage P. If weeks are used as the time unit, it can be expressed as (P=N1, N2, N3). Taking the i-th week as an example, if it is in the fruiting period, then T _Mi =T _N3 .

(302) Calculate the daily average temperature within seven days at 00:00 every day.

Divide a week into seven days, input the seven-day weather forecast combined with the crop physiological model and the greenhouse energy consumption model, and substitute into the corresponding economic benefit model, with the goal of maximizing the economic benefits of greenhouse regulation, and rolling the cumulative temperature of the next week Assign it to each day, and obtain the daily average temperature T _D1 , T _D2 ... T _D7 within a week. The rolling optimization method eliminates the boundary between weeks in the hierarchical process and solves the problem that a week spans physiological stages, which is also one of the innovations of the present invention. Since the temperature factor is mainly considered in this step, the daily economic benefit is expressed as the difference between the economic income of crops and the cost of heating energy consumption, and the performance function J1 is _:

In the formula, q _tom [yuan/mg] is the unit price of agricultural products, η _DMFM is the conversion factor from fruit dry weight to fruit fresh weight, DM _Har [mg/m ² ] is the harvested fruit dry matter yield, q _heat [yuan/J] is the unit price of heating energy, and Q _heat [J/m ² ] is heating energy consumption.

The constraints used in the rolling optimization include seven-day cumulative temperature conditions and indoor temperature upper and lower limit conditions.

The main purpose of the optimization of this layer is to distribute the average temperature of the week (the optimization of this layer takes the week as a cycle) to each day of the week with the goal of saving energy according to the changes of the external weather. The goal is to achieve the highest economic benefit, and at the same time, there are constraints on the temperature tolerance, and the daily average temperature is optimized by rolling and iteratively. Combined with the above analysis, the optimization problem of this layer can be summarized as a nonlinear maximum optimization problem with constraints.

For the convenience of explaining the problem, taking the seven days of the first week and the seven days of the second week as examples, assuming that the outdoor weather in the next week can always be accurately forecasted, and considering the requirement of meeting the accumulated temperature, then the optimal day of each day in these two weeks The average temperature is obtained by rolling optimization as follows, and the details of the iterative process are described as follows:

At 00:00 am on the first day, the first iterative optimization of the first week was performed to determine the average temperature on the first day:

T _D ＝[T _D1 ,T _D2 ,T _D3 ,T _D4 ,T _D5 ,T _D6 ,T _D7 ]

W _D ＝[W _D1 ,W _D2 ,W _D3 ,W _D4 ,W _D5 ,W _D6 ,W _D7 ]

Among them, J ₁ represents the objective function of economic benefits, T _min and T _max are the allowable upper and lower limits of indoor temperature respectively, T _M1 is the average temperature in the first week, W _D is the weather forecast vector for one week outdoors, which is used as the fixed input of J ₁ . Considering that temperature control needs to meet the crop's demand for accumulated temperature, the daily temperature should be equal to the weekly accumulated temperature demand. Then the optimal daily average temperature on the 1st day is T _D1,opt , but T _D2,opt ~ T _D7,opt is not directly set as the optimal daily average temperature from the 2nd to the 7th day, because in fact, the There must be a certain difference between the final actual daily average temperature from the 2nd to the 7th day and the optimal daily average temperature obtained in the first iteration, and this difference must be compensated in the subsequent iterative optimization.

At 00:00 am on the second day, the second iterative optimization in the first week was performed to determine the average temperature on the second day:

T′ _D ＝[T′ _D2 , T′ _D3 , T′ _D4 , T′ _D5 , T′ _D6 , T′ _D7 , T′ _D8 ]

W _D ＝[W _D2 ,W _D3 ,W _D4 ,W _D5 ,W _D6 ,W _D7 ,W _D8 ]

Among them, T′ _D,opt is the daily average temperature optimized in the second iteration, T _D1 is the actual average temperature on the first day, T _M1 is the average temperature in the first week, and T _M2 is the average temperature in the second week. The influence of the boundary is optimized in a rolling manner, and the rolling is reflected in the constraints of the average temperature of the next seven days, that is, the sum of the optimization results of the average temperature of the 2nd to 8th days and the sum of the temperatures of the 2nd to 8th days minus the actual temperature of the first day. Then the optimal daily average temperature of the next day can be taken as T′ _D2,opt .

By analogy, at 00:00 am on the 7th day, the 7th iterative optimization of the 1st week is performed to determine the average temperature on the 7th day:

W _D ＝[W _D7 ,W _D8 ,W _D9 ,W _D10 ,W _D11 ,W _D12 ,W _D13 ]

Among them, W _D is the weather forecast vector for one week outside, T _D1 -T _D6 is the actual average temperature of the first to sixth days, T _M1 is the average temperature of the first week, and T _M2 is the average temperature of the second week. Considering that temperature control needs to meet the crop's demand for accumulated temperature, the daily temperature should be equal to the weekly accumulated temperature demand. Then the optimal daily mean temperature on day 7 is but It is not directly set as the optimal daily average temperature from the 8th to the 13th day.

By analogy, in each iteration process, the constraint conditions must accumulate the average temperature of the next 7 days to ensure that the total accumulated temperature meets the needs of crop growth. The optimal average temperature of each day is taken from the first variable of each iteration. In summary, taking the first week as an example, the entire iterative process is expressed as follows:

By analogy, at 00:00 am on the 8th day, the first iterative optimization of the second week is performed to determine the average temperature on the 8th day:

T _D ＝[T _D8 ,T _D9 ,T _D10 ,T _D11 ,T _D12 ,T _D13 ,T _D14 ]

W _D ＝[W _D8 ,W _D9 ,W _D10 ,W _D11 ,W _D12 ,W _D13 ,W _D14 ]

Among them, T _M2 is the average temperature of the second week, and W _D is the weather forecast vector of the outdoor week. Considering that temperature control needs to meet the crop's demand for accumulated temperature, the daily temperature should be equal to the weekly accumulated temperature demand. Then the optimal daily average temperature on the 8th day is T _D8,opt , but T _D9,opt ~ T _D14,opt is not directly set as the optimal daily average temperature from the 9th to the 14th day.

(303) Take the calculation result of the first day's average temperature obtained in step (302) as the constraint condition for optimization, and calculate the temperature setting value from this moment to the next full point at each hour.

Since photosynthesis is closely related to the greenhouse environment during the day, and the daily light changes are relatively severe, it can be estimated by combining the cloudy and sunny conditions of each hour. Input the weather forecast in units of hours combined with the crop physiological model and the greenhouse energy consumption model, with the goal of maximizing the economic benefits of greenhouse regulation, rolling optimization to obtain the daily average temperature constraints obtained in step (302) for each hour. Average temperature T _H1 , T _H2 ... T _H24 . The method of rolling optimization is adopted, and the deviation is corrected with the actual temperature of the past time of the day, which can ensure that the daily average temperature is reached, which is also one of the innovation points of the present invention. Since the temperature factor is mainly considered in this step, the daily economic benefit is expressed as the difference between the economic income of crops and the cost of heating energy consumption, and the performance function J2 is _:

In the formula, [yuan/mg] q _tom is the unit price of agricultural products, η _DMFM is the conversion factor from fruit dry weight to fruit fresh weight, DM _Har [mg/m ² ] is the harvested fruit dry matter yield, q _heat [yuan/J] is the unit price of heating energy, and Q _heat [J/m ² ] is heating energy consumption. It is worth noting that because there is no natural light at night, crops do not perform photosynthesis and do not produce photosynthetic products.

Constraints used in rolling optimization include daily cumulative temperature conditions, indoor temperature upper and lower limit conditions, daytime average temperature conditions and adjacent hour temperature difference upper limit conditions.

For the convenience of explaining the problem, take the temperature optimization of a certain day when the outdoor weather is cold wave as an example, assuming that the outdoor weather of this day in the future can always be accurately predicted, then the optimal average temperature in the greenhouse per hour in this day is rolling and optimized as follows Method to get:

At 00:00 am on the first day, the first iterative optimization on the first day is carried out to determine the temperature setting value from 00:00 to 01:00:

T _H ＝[T _H1 ,T _H2 ,T _H3 ... T _H24 ]

W _H ＝[W _H1 ,W _H2 ,W _H3 ...W _H24 ]

and T _min ≤T _Hk ≤T _max

and|T _Hi -T _Hi-1 |≤m(i＝2,3,4,...24)

Among them, J ₂ represents the economic benefit objective function, T _min and T _max are the allowable upper and lower limits of indoor temperature respectively, T _D1,opt is the average temperature of the first day optimized by the previous layer, Indicates the average daytime temperature, n indicates the optimum temperature difference between day and night for crops, and m is the upper limit of temperature difference between adjacent hours. Then the optimal hourly average temperature optimized in the first time window is T _H1,opt , but _TH2,opt ~ _TH24,opt is not considered as the optimal hourly average temperature in the second to 24th time window, because the actual In general, there must be a certain difference between the final actual average temperature from the 2nd to the 24th hour and the optimal hourly average temperature obtained in the first iteration, and this difference must be compensated in the subsequent iterative optimization.

At 01:00 am on the first day, the second iterative optimization on the first day is carried out to determine the temperature setting value from 01:00 to 02:00:

T′ _H ＝[T′ _H2 , T′ _H3 , T′ _H4 ... T′ _H25 ]

W _H ＝[W _H2 ,W _H3 ,W _H4 ...W _H25 ]

and T _min ≤T′ _H ≤T _max

and|T′ _Hi -T′ _Hi-1 |≤m(i=3,4...25)

Among them, T _H1,real is the actual hourly average temperature of the first hour, T _D1,opt and T _D2,opt are the average temperature of the first day and the second day respectively obtained by the optimization of the previous layer, in order to overcome the influence of the boundary, The optimization is carried out in a rolling manner, and the rolling is reflected in the constraints of the average temperature of the next 24 hours, that is, the sum of the optimization results of the average temperature in the 2nd to 25th hours and the sum of the temperatures in the 1st to 25th hours minus the actual temperature in the first hour. Then the optimal hourly average temperature in the second time window can be taken as T′ _H2,opt .

T′ _H ＝[T′ _H2 , T′ _H3 , T′ _H4 ... T′ _H25 ]

W _H ＝[W _H2 ,W _H3 ,W _H4 ...W _H25 ]

and T _min ≤T′ _H ≤T _max

and|T′ _Hi -T′ _Hi-1 |≤m(i=3,4...25)

By analogy, in each iteration, the constraint conditions must accumulate the real hourly average temperature to ensure the daily average temperature requirement, and at the same time meet the temperature difference between day and night. The optimal average temperature per hour is taken from the first variable of each iteration. At 23:00 on the first day, carry out the 24th iterative optimization on the first day to determine the temperature setting value from 23:00 to 00:00:

T” _H ＝[T” _H24 ,T” _H25 ,T” _H26 ...T” _H47 ]

W _H ＝[W _H24 ,W _H3 ,W _H4 ...W _H47 ]

and T _min ≤ T” _H ≤ T _max

and|T” _Hi -T” _Hi-1 |≤m(i＝25,26...48)

Among them, T _{H1, real} to T _{H23, real} are the actual hourly average temperatures from the 1st to 23rd hour respectively, and T _{D1, opt} and T _{D2, opt} are the average temperature of the first day and the second day obtained from the optimization of the previous layer. Temperature, in order to overcome the influence of the boundary, optimize in a rolling manner, and the rolling is reflected in the constraint of the average temperature of the next 24 hours, that is, the sum of the optimization results of the average temperature of the 24th to 47th hour, the sum of the temperature of the 1st to 47th hour minus the first 1- 23 hours actual temperature. Then the temperature setting value for the 24th hour can be taken as T” _P24,opt .

(304) Calculating the carbon dioxide concentration setting value of each control step within one hour at each hour.

In each hour, because the temperature change rate in the control step (15 minutes is taken as an example in this embodiment) is relatively slow, the temperature setting value does not change, but the change rate of the carbon dioxide concentration is relatively fast. Taking the temperature setting value obtained in step (303) and the light in the weather forecast as the conditions, taking the indoor carbon dioxide concentration setting value as the optimization variable, and aiming at maximizing the total economic benefits of the control step, solve the four most economic benefits of greenhouse regulation. CO _2t1 , CO _2t2 , CO _2t3 , CO _2t4 are set in each carbon dioxide concentration step, and the carbon dioxide constraint does not include the upper equation constraint, so it is not necessary to use rolling optimization. Since this step only considers carbon dioxide control, the economic benefit function corresponding to the control step is the difference between the economic income of crops and the cost of carbon dioxide, and the performance function _J3 is:

In the formula, q _tom [yuan/mg] is the unit price of agricultural products, the conversion factor of η _DMFM fruit dry weight to fruit fresh weight, η _PDM is the conversion factor of photosynthetic products into dry matter, and P [mg/m ² ] is the control period total photosynthetic output, [yuan kg ^-1 ] is the unit of carbon dioxide, [kg/m ² ] is the amount of carbon dioxide released. It is worth noting that because there is no natural light at night, crops do not perform photosynthesis and do not produce photosynthetic products.

(4) Obtain real-time values of indoor environmental factors, combine with the environmental setting values in step (3), and control the corresponding actuators in the greenhouse based on the principle of regulating the coordination of environmental factors and the mutual coordination of regulation means, and the specific execution mechanism control program flow As shown in Figure 9, the specific description is as follows:

There are many implementing agencies in the greenhouse, and the energy consumption of different control methods is not the same. For example, when the greenhouse needs to be cooled, simple ventilation is used to cool down compared to wet curtain water pumps and fans. Although the cooling effect is limited, the energy cost is lower.

The basic principles of greenhouse control are:

a) The protection system should be sensitive and protect the greenhouse facilities in case of heavy rain and strong wind;

b) Prioritize the use of actuators with lower energy consumption for control, such as ventilation;

c) In spring and autumn, pay attention to moderate heat preservation, mainly through ventilation and curtains to regulate the greenhouse environment;

d) Pay attention to heat preservation in winter to avoid low temperature stress;

e) Pay attention to cooling and shading in summer to avoid high temperature stress and sunburn;

f) Pay attention to dehumidification to reduce pests and avoid plant rot;

Existing studies have shown that temperature plays a vital role in both the growth and development of crops and the control of energy consumption in greenhouses. Therefore the present invention divides multiple controlled factors in the greenhouse into major and minor categories according to their importance, and the minor factors (such as humidity, light, etc.) are all managed to coordinate with the main factors (such as temperature) to find out the coordination relationship function, thereby Transform complex multi-factor control into multi-factor coordinated control based on temperature single factor, supplemented by feed-forward and feedback control to eliminate certain uncertainties brought about by "coordination", and solve the problem of severe coupling of multiple factors in the greenhouse environment , to achieve the purpose of multi-factor control. A weighted linear function T of the combination of indoor and outdoor parameters and manual setting values determines the action of each control method. The formula is as follows:

T(m _co2 ,m _T ,m _R ,m _H )＝α×F(m _co2set ,m _Tset ,m _Rset ,m _Hset )+β·G(m _co2in ,m _Tin, m _Rin ,m _Hin )+λ ·H(m _Tout ,m _Rout ,m _Hout ,Fv,P _rain )

Among them, T is the specific control method, F is the parameter value function set manually, G is the outdoor environment parameter, H is the indoor environment parameter; α, β, λ are the corresponding weights; m _co2 is the amount of carbon dioxide released, m _T is the target temperature, m _R is the target radiation amount of light, m _H is the target humidity, m _co2set is the set value of indoor carbon dioxide concentration, m _Tset is the set value of indoor temperature, m _Rset is the set value of indoor light radiation, m _Hset is indoor humidity setting value, m _co2in is indoor carbon dioxide concentration, m _Tin is indoor temperature, m _Rin is indoor light radiation, m _Hin is indoor humidity, m _co2out is outdoor carbon dioxide concentration, m _Tout is outdoor temperature, m _Rout is Outdoor light radiation, m _Hout is the outdoor humidity, Fv is the outdoor wind speed, P _rain is the outdoor rainfall.

(401) temperature control

The most important and complex factor in the greenhouse, other factors directly or indirectly affect the temperature, and the same temperature also affects them (except for light). For the consideration of energy saving, when cooling is required, ventilation is preferred to control the greenhouse. If the ventilation control results cannot meet the requirements, refrigeration (wet curtain water pump and fan or spray) is used to cool down. When the temperature needs to be raised, the heat preservation net is preferred for heat preservation. When the heat preservation result still cannot meet the requirements, the greenhouse heating is turned on. Because heating during ventilation will bring a lot of heat waste, it is generally not suitable to heat the greenhouse during ventilation. Target temperature:

mT＝J(m _co2 ,m _R ,m _H ,K,K',Fv)

In the case of ventilation, in order not to increase the difficulty of control, a relatively simple coordination relationship can be considered, but it must be within the allowable control effect range. Therefore, after simplification of gardening experience, the influence of other environmental factors on ventilation temperature is shown in Figure 5 and Figure 6, and the corresponding Figures 7 and 8 show the correction process of humidity and light intensity on ventilation temperature.

In the case of heating (cooling), the indoor temperature setting value is obtained according to the layered rolling optimization step in step (3), and the above-mentioned temperature setting value is achieved by regulating the actuator of the greenhouse heating (refrigeration).

Target temperature mT: mT _(L) ≤ mT ≤ mT _(U) , 24 hours a day, the target temperature is obtained by step (3). Taking tomato as an example, the maximum temperature limit during the day is 35°C, the suitable temperature is 18-25°C, the minimum temperature limit at night is 5°C, and the suitable temperature is 8-13°C.

(402) Humidity Control

Temperature and humidity are two variables with strong coupling. It can be known from practical experience that temperature has a great influence on humidity. The higher the temperature, the lower the humidity will be. According to the measurement, the relative humidity will drop by 2% for every 1°C increase in temperature. -3%. The influence of humidity on temperature is small and can be ignored, and the change of humidity is much slower than that of temperature, so it can be completely decoupled by compensation. With temperature compensation for humidity, both temperature and temperature variables can be treated as univariate. Compared with temperature control, humidity control is relatively simple. We use some equivalent experience knowledge to realize the transformation of control objectives, and complete it by controlling and coordinating the opening of skylights and side windows, and starting and stopping internal spraying and uniform fans.

Target Humidity:

mH＝J(m _T ,m _co2 ,m _R ,K,K',Fv)

mH _(L) ≤ mH ≤ mH _(U) should be satisfied, and it varies with the growth period of the crop. The general rule is: 55% at the seedling stage, 70% at the growth stage, and 80% at the fruit stage.

(403) Lighting Control

Lighting control is a relatively independent link. In terms of control means, light control is reflected through the sunshade net. The photosynthesis of crops is very important. Sufficient light time must be guaranteed in a day to ensure the normal growth of crops, but too strong light will easily cause damage to the surface of crops, so appropriate measures must be taken.

The greater the light intensity, the faster the temperature in the greenhouse will rise. The impact of light on other environmental factors is unilateral. No matter how other environmental factors change, the light intensity will not be affected in any way. Therefore, the lighting control is relatively independent. The illumination target mR should meet:

mR _(L) ≤mR≤mR _(U) .

(404) Carbon Dioxide Control

Carbon dioxide is essential for the growth of crops. Carbon dioxide can increase the yield of crops, increase the fruit, and reduce the occurrence of diseases and insect pests. However, too much carbon dioxide will inhibit the growth of crops and increase the indoor temperature. Therefore, we must grasp the amount of carbon dioxide. It can be applied in the early and middle stages of vegetable growth, and the effect of application in the period of rapid fruit expansion is the best.

The supply of carbon dioxide mainly depends on the strength of photosynthesis, and the strength of photosynthesis is related to light and temperature. In turn, carbon dioxide affects the temperature in the room. During the day, the stronger the light, the higher the photosynthesis and surface temperature of the crops, and the crops will absorb more carbon dioxide; similarly, carbon dioxide will cause the "greenhouse effect" and increase the indoor temperature. At night, crops do not carry out photosynthesis, and the application of carbon dioxide can be ignored. In step (3), the above-mentioned influences are comprehensively considered to obtain the carbon dioxide setting value of each control step with the best income increasing effect.

In the ventilation state, the artificially supplemented carbon dioxide will diffuse into the air, which not only fails to increase the indoor carbon dioxide concentration, but may also aggravate the global greenhouse effect. In addition, the cost of artificial carbon dioxide is high, so it is not suitable to supplement carbon dioxide during ventilation.

During the day, when the light intensity is high, the photosynthesis of crops is enhanced. At this time, increase the concentration of carbon dioxide, but there must be a certain upper limit; when the light intensity is low, reduce the concentration of carbon dioxide, but also keep it above a certain lower limit. :

m _co2(L) ≤ m _co2 ≤ m _co2(U) .

(405) Shade net control

Setting parameters: the threshold value mR(U) of sunshade net deployment, mR(U)=200W/m ² in seedling stage, mR(U)=300W/m ² in seedling stage, mR(U)=800W/m ² in fruit stage. Sunshade net roll-up protection: strong wind protection Fv(U), rainstorm protection Prain(U). When the light R≥mR(U), unfold the sunshade net, otherwise close it; when the wind speed Fv≥Fv(U) or the rainfall Prain≥Prain(U) roll up the sunshade net. Among them, R is the measured value of light, Fv is the measured value of wind speed, and the measured value of Prain rainfall.

(406) Cooling (spray, wet curtain) control

The control of the cooling system is similar to that of the heating system. The temperature setting value obtained in step (3) is the opening threshold of the cooling system, and the dead zone ΔmT(U) is set.

When the indoor temperature T≥mT(U), the cooling system is turned on, and the cooling system is turned off when T≤mT(U)-ΔmT(U). Usually in summer T≤mT(U)-ΔmT(U).

(407) Sunroof Control

The sunroof control process is shown in Figure 10, setting parameters: frosting temperature T' = 2 °C, ventilation temperature T" = T _i + ΔmT _Hi + ΔmT _Ri (the calculation process of ventilation temperature T" is shown in Figure 7 and Figure 8 As shown, mTi is the target temperature in the i-th period, ΔmT _Hi is the correction value of the humidity to the ventilation temperature, and ΔmT _Ri is the correction value of the light radiation intensity to the ventilation temperature). There are 3 levels of skylight opening: K1<K2<K3, K0=0, K1=10%, K2=50%, K3=100%.

When the outdoor temperature T _out ≤ T' or Fv ≥ Fv (U), do not open the window (that is, the skylight opening K = 0);

When the outdoor temperature satisfies T'<T _out ≤mT(L), K=K1=10%;

When the outdoor temperature satisfies mT(L)<T _out ≤T", K=K2=50%;

When the outdoor temperature satisfies T _out >T", the skylight is fully opened K=K3=100%.

Protection algorithm: when the rainfall Prain≥Prain(U) or the wind speed Fv≥Fv(U), close the skylight (that is, K=0).

(408) Side window control

Set the parameter side window to adjust the temperature range ΔT".

When T _out ≤ T", the side window does not open, K'=0;

When T _out >T", the side window opening K' has two levels: K ₁ '(50%), K ₂ '(100).

details as follows:

K'=K ₁ '(50%) when T"<T _out ≤T"+ΔT"

K'=K' ₂ (100%) when T _out >T"+ΔT"

Protection algorithm: when the rainfall Prain≥Prain(U) or the wind speed Fv≥Fv(U), close the side windows (that is, K'=0).

(409) Insulation net control

Setting parameters: the upper and lower bounds M(L) and M(U) of the month when the thermal insulation net is deployed, and the upper and lower bounds of time T(L) and T(U) (usually winter nights). Insulation net roll-up protection: strong wind protection Fv(U), rainstorm protection Prain(U). The current time is in the open month M(L)≤M≤M(U) and the open period T(L)≤T≤T(U), unfold the insulation net, otherwise close; when the wind speed Fv≥Fv(U) or the rainfall Prain ≥Prain(U) rolls with insulation net.

(410) Heating system control

In the case of greenhouse heating, the method of layered optimization is used to obtain the optimal numerical solution, and the temperature setting value is dynamically obtained under the condition of ensuring the accumulated temperature demand, so as to reduce the energy consumption of the greenhouse.

The temperature setting value obtained in step (3) is the working threshold of the heating system, and the dead zone ΔmT(L) is set.

When the indoor temperature T≤mT(L), turn on the heating system until T≥mT(L)+ΔmT(L), usually ΔmT(L)=1°C in winter.

The preferred specific embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make many modifications and changes according to the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning or limited experiments on the basis of the prior art shall be within the scope of protection defined by the claims.

Claims (10)

1. A greenhouse environment multi-factor coordinated energy-saving optimization control method, characterized in that, comprising the following steps: 1) Set the expected daily average temperature of crops in each growth period, and obtain weather forecast data for the next seven days; 2) Estimate the status of the greenhouse ventilation system; 3) According to step 1) and step 2), adopt multi-factor coordination algorithm to set the setting values of each environmental factor in the greenhouse, and the environmental factor includes temperature, humidity, light radiation intensity and carbon dioxide concentration; 4) Obtain the real-time value of the environmental factor, and adjust the corresponding actuator in the greenhouse according to the set value of the environmental factor. 2. The greenhouse environment multi-factor coordination energy-saving optimization control method according to claim 1, characterized in that, in the step 3), the specific process for setting the temperature is: A1) Determine the weekly average temperature according to the growth period of the weekly crops; A2) According to the expected daily average temperature and weather forecast data, the rolling optimization method is used to calculate the optimal daily average temperature for the next seven days. The frequency of rolling optimization is once _a day. The performance function J1 used in rolling optimization is: <mrow> <msub> <mi>J</mi> <mn>1</mn> </msub> <mo>=</mo> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>7</mn> </munderover> <msub> <mi>q</mi> <mrow> <mi>t</mi> <mi>o</mi> <mi>m</mi> </mrow> </msub> <msub> <mi>&amp;eta;</mi> <mrow> <mi>D</mi> <mi>M</mi> <mi>F</mi> <mi>M</mi> </mrow> </msub> <msub> <mi>DM</mi> <mrow> <mi>H</mi> <mi>a</mi> <mi>r</mi> </mrow> </msub> <mrow> <mo>(</mo> <msub> <mi>T</mi> <mrow> <mi>D</mi> <mi>i</mi> </mrow> </msub> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>q</mi> <mrow> <mi>h</mi> <mi>e</mi> <mi>a</mi> <mi>t</mi> </mrow> </msub> <msub> <mi>Q</mi> <mrow> <mi>h</mi> <mi>e</mi> <mi>a</mi> <mi>t</mi> </mrow> </msub> <mrow> <mo>(</mo> <msub> <mi>T</mi> <mrow> <mi>D</mi> <mi>i</mi> </mrow> </msub> <mo>)</mo> </mrow> </mrow> In the formula, q _tom η _DMFM DM _Har (T _Di ) represents the economic income produced by the crop when the daily average temperature of the i day is T _Di , q _tom represents the unit price of the crop, and η _DMFM represents the conversion factor from fruit dry weight to fruit fresh weight , DM _Har represents the harvested fruit dry matter yield, q _heat Q _heat (T _Di ) represents the heating energy consumption cost when the i-th day’s average temperature is T _Di , q _heat represents the unit price of heating energy, Q _heat represents the heating energy consumption, The constraints used in the rolling optimization include seven-day cumulative temperature conditions and indoor temperature upper and lower limit conditions; A3) The rolling optimization method is used to calculate the hourly average temperature of the day under the optimal daily average temperature constraint, the frequency of rolling optimization is once per hour, and the performance function J2 adopted during rolling optimization is _: <mrow> <msub> <mi>J</mi> <mn>2</mn> </msub> <mo>=</mo> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>j</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>24</mn> </munderover> <msub> <mi>q</mi> <mrow> <mi>t</mi> <mi>o</mi> <mi>m</mi> </mrow> </msub> <msub> <mi>&amp;eta;</mi> <mrow> <mi>D</mi> <mi>M</mi> <mi>F</mi> <mi>M</mi> </mrow> </msub> <msub> <mi>DM</mi> <mrow> <mi>H</mi> <mi>a</mi> <mi>r</mi> </mrow> </msub> <mrow> <mo>(</mo> <msub> <mi>T</mi> <mrow> <mi>H</mi> <mi>j</mi> </mrow> </msub> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>q</mi> <mrow> <mi>h</mi> <mi>e</mi> <mi>a</mi> <mi>t</mi> </mrow> </msub> <msub> <mi>Q</mi> <mrow> <mi>h</mi> <mi>e</mi> <mi>a</mi> <mi>t</mi> </mrow> </msub> <mrow> <mo>(</mo> <msub> <mi>T</mi> <mrow> <mi>H</mi> <mi>j</mi> </mrow> </msub> <mo>)</mo> </mrow> </mrow> In the formula, T _Hj represents the hourly average temperature of the jth hour; Constraints used in rolling optimization include daily cumulative temperature conditions, indoor temperature upper and lower limit conditions, daytime average temperature conditions and adjacent hour temperature difference upper limit conditions. 3. The greenhouse environment multi-factor coordinated energy-saving optimization control method according to claim 2, characterized in that, the growth period of the crop includes seedling stage, growth stage and fruit stage. 4. The greenhouse environment multi-factor coordinated energy-saving optimization control method according to claim 2, characterized in that, when obtaining the economic income produced by the crops, the crop yield or fruit dry matter are equivalently distributed to each growth stage of crop growth. developmental stage. 5. The greenhouse environment multi-factor coordination energy-saving optimization control method according to claim 1, characterized in that, in the step 3), the specific process for setting the carbon dioxide concentration is: In each hour, based on the current temperature setting value and the light data in the weather forecast data as conditions, the carbon dioxide concentration setting value as the optimization variable, and the goal of maximizing the sum of economic benefits of the control step, optimize and calculate To obtain the carbon dioxide concentration setting value of each control step, the sum of the economic benefits adopted in the optimization process, that is, the performance function J ₃ is: <mrow> <msub> <mi>J</mi> <mn>3</mn> </msub> <mo>=</mo> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>s</mi> </munderover> <msub> <mi>q</mi> <mrow> <mi>t</mi> <mi>o</mi> <mi>m</mi> </mrow> </msub> <msub> <mi>&amp;eta;</mi> <mrow> <mi>D</mi> <mi>M</mi> <mi>F</mi> <mi>M</mi> </mrow> </msub> <msub> <mi>&amp;eta;</mi> <mrow> <mi>P</mi> <mi>D</mi> <mi>M</mi> </mrow> </msub> <mi>P</mi> <mrow> <mo>(</mo> <msub> <mi>CO</mi> <mrow> <mn>2</mn> <mo>,</mo> <mi>k</mi> </mrow> </msub> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>q</mi> <mrow> <msub> <mi>CO</mi> <mn>2</mn> </msub> </mrow> </msub> <msub> <mi>m</mi> <mrow> <msub> <mi>CO</mi> <mn>2</mn> </msub> </mrow> </msub> <mrow> <mo>(</mo> <msub> <mi>CO</mi> <mrow> <mn>2</mn> <mo>,</mo> <mi>k</mi> </mrow> </msub> <mo>)</mo> </mrow> </mrow> In the formula, CO _2,k represents the set value of carbon dioxide concentration at the kth control step, q _tom represents the unit price of agricultural products, η _DMFM represents the conversion factor from fruit dry weight to fruit fresh weight, η _PDM represents the conversion factor of photosynthetic products into dry matter Conversion factor, P represents the total photosynthetic output in the control period, Indicates the carbon dioxide unit, Indicates the amount of carbon dioxide released, and s indicates the total number of control steps in one hour. 6. The greenhouse environment multi-factor coordinated energy-saving optimization control method according to claim 5, wherein the control step is 15 minutes. 7. The greenhouse environment multi-factor coordinated energy-saving optimization control method according to claim 1, characterized in that, in said step 2), the estimated state of the greenhouse ventilation system is specifically: The temperature value in the weather forecast data is used as the outdoor temperature, the outdoor temperature is compared with the frosting temperature and the ventilation temperature, and the opening degree of the ventilation system is obtained according to the comparison result. 8. The greenhouse environment multi-factor coordinated energy-saving optimization control method according to claim 1, characterized in that, in the step 4), when regulating and controlling the corresponding actuators in the greenhouse, the mutual coordination between the regulating and controlling greenhouse environmental factors and the mutual regulation means Coordination is the principle. 9. The greenhouse environment multi-factor coordinated energy-saving optimization control method according to claim 1, characterized in that, in the step 4), when the greenhouse is regulated, it specifically includes temperature control, humidity control, illumination control, carbon dioxide control and Ventilation control. 10. The greenhouse environment multi-factor coordinated energy-saving optimization control method according to claim 1, characterized in that, in said step 4), when regulating and controlling the greenhouse, a weighted linear function T is used to determine the actions of each control means, and said The expression of the weighted linear function T is: T(m _co2 ,m _T ,m _R ,m _H )＝α×F(m _co2set ,m _Tset ,m _Rset ,m _Hset )+β·G(m _co2in ,m _Tin ,m _Rin ,m _Hin ) +λ·H(m _Tout ,m _Rout ,m _Hout ,Fv,P _rain ) In the formula, T represents the specific control means, F represents the parameter value function manually set, G represents the indoor environment parameters, H represents the outdoor environment parameters, α, β, λ represent the corresponding weights respectively; m _co2 represents the amount of carbon dioxide released, m _T represents the target temperature, m _R represents the target radiation amount of light, m _H represents the target humidity, m _co2set represents the set value of indoor carbon dioxide concentration, m _Tset represents the set value of indoor temperature, m _Rset represents the set value of indoor light radiation, m _Hset indicates indoor humidity setting value, m _co2in indicates indoor carbon dioxide concentration, m _Tin indicates indoor temperature, m _Rin indicates indoor light radiation, m _Hin indicates indoor humidity, m _co2out indicates outdoor carbon dioxide concentration, m _Tout indicates outdoor temperature, m _Rout Indicates the amount of outdoor light radiation, m _Hout indicates the outdoor humidity, Fv indicates the outdoor wind speed, and P _rain indicates the outdoor rainfall.